The idea of AI was developed in 1950s. The first attempt of artificial intelligence was in 1950 by a researcher, Alan Turing. He published “Computing Machinery and Intelligence,” where in he proposed an imitation game dubbed the “Turing test.”
From there on we have seen astounding uses of this innovation – from the 1st human-like robot, Honda’s ASIMO to IBM’s Watson, the supercomputer that went ahead to crush the two biggest champions in the Television Quiz program’s history. Yet, we have not seen a wide utilization of artificial intelligence or machine learning in a business setting.
If we look at today’s cyberthreats, the introduction and implementation of Artificial Intelligence is quite beneficial for cybersecurity professionals. AI systems comes with a package which consists of number of functionalities which would benefit cybersecurity professionals to better safeguard the business.
These AI algorithms uses Machine Learning (ML) in order to understand and carry out any set action which is required.
Every day the malwares & cyber-attacks are becoming more sophisticated and smart for it to be detected by normal cybersecurity protocols. Eg: Baiting technique used by malware since every software has a set memory in a computer .This bait is known as a folder bomb which is an infinite looped folder used by the malware. When the antivirus(AV) executes the bait folder it tries to search the folder thus forcing the anti-virus to crash as it has used all its memory to open the file. Once the AV crashes the malware executes its payload thus spreads the infection to the computer. This technique was used quite earlier. Now antiviruses can identify it. If a AV finds a folder bomb it immediately send the threatening material to the quarantine section.
Therefore computer security depends on ML to use information from past computer-invasions in order to cope up with new/similar risk without wasting time. AI can be used to find and stop simple threats and attacks. Since a simple attack can most probably have a simple answer, AI can solve the situation by itself. It depends upon wise automatic trigger system used in AI which raises hazard warnings that people might not have found. It also lessens the need for hiring more staff and results in better effectiveness being automated or simulated.
One more way an AI system can be useful is by ranking attacks based on their level. Thus it prioratises where and what defenses are to be implemented on the basis of the strength of the attack.
However, it is said that 52% of the Cybersecurity professionals are not happy with AI systems in cybersecurity as they feel that it does not do an accurate job. What needs to be clearly understood is that when the machine learning principles is extensively applied and over time it understands and adapts thus providing necessary results.
Yet, disappointingly, there will be human dependency even after AI application in cybersecurity, We can see that in real life the weather forecast prediction happens via AI which results in us cursing the Bureau of meteorology for predicting wrong weather conditions. With constant improvements the AI is seemed to perform better. Eg:By updating the new hurricane patterns it helps the meteorology department to guage hurricane formation and rank them in order to find its strength.
As AI gets better and effective there are agents who love to confuse the existing functionality of the working AI model. I hope most of you know about the security issue in Microsoft word especially in the macro’s which is patched. Recently I stumbled across a opensource software which has an embedded word template. When the attachment is sent to the intended target the anti-virus thinks it is an word document and the Microsoft word thinks it is one of its template thus it enables the malware to get the credentials of the Microsoft accounts connected with the Microsoft office tools. This was a great thing to explore. I am trying ways to patch this up but also could not help reflecting how clever these exploiters are. If this vulnerability falls in wrong hands, well ………..YOU HAVE BEEN HACKED!…………..